# robots.txt — Nefro-projekt Slovensko
# https://nefro.polascin.net/
# Aktualizované: 2026-09-25

# ── Všeobecné pravidlá pre všetkých robotov ────────────────────────────────
User-agent: *
Allow: /

# ── Preferencie použitia obsahu AI systémami ───────────────────────────────
# IETF aipref (draft-ietf-aipref-attach).
#   train-ai=n — obsah sa nesmie použiť na trénovanie generatívnych modelov
#   search=y   — indexovanie a odkazovanie je vítané
# Tú istú výhradu nesú HTTP hlavičky (Content-Usage, tdm-reservation),
# /.well-known/tdmrep.json a Podmienky používania (/terms.php#tdm).
Content-Usage: train-ai=n, search=y

# Administratívne a vývojové stránky
Disallow: /admin.php
Disallow: /admin_articles.php
Disallow: /admin_discussion.php
Disallow: /admin_newsletter.php
Disallow: /admin_legal_notice.php
Disallow: /admin_trials.php
Disallow: /admin_drugs.php
Disallow: /admin_drugs_review.php
Disallow: /admin_providers.php
Disallow: /admin_reauth.php
Disallow: /admin_users_notice.php
Disallow: /query_article.php
Disallow: /setup_db.php
Disallow: /archive_cleanup.php
Disallow: /newsletter_worker.php
Disallow: /newsletter_weekly_digest.php
Disallow: /resend_verification.php
Disallow: /verify_email.php
Disallow: /newsletter_verify_sub.php
Disallow: /newsletter_unsubscribe.php
Disallow: /newsletter_subscribe.php
Disallow: /logout.php
Disallow: /avatar_upload.php
Disallow: /calculator_result_print.php
Disallow: /search.php?*
Disallow: /discussion.php
Disallow: /profile.php
Disallow: /login.php
Disallow: /register.php
Disallow: /2fa_setup.php
Disallow: /2fa_verify.php
Disallow: /email_verification.php
Disallow: /mobile_verification.php
Disallow: /forgot_password.php
Disallow: /reset_password.php
Disallow: /confirm_account_deletion.php
Disallow: /session_keepalive.php
Disallow: /calculator_history.php

# Honeypot — pasca na crawlery, ktoré tento súbor ignorujú (pozri bot_trap.php).
# Kto sem napriek tomuto zákazu príde, dostane dočasný ban.
Disallow: /bot_trap.php

# Privátne adresáre
Disallow: /private/
Disallow: /uploads/avatars/archive/
Disallow: /hooks/
Disallow: /scratch/
Disallow: /.git/

# Vývojové súbory
Disallow: /*.py
Disallow: /*.ps1

# Crawl delay pre šetrenie servera (nízka hodnota — malý web chce byť indexovaný;
# Google Crawl-delay ignoruje, Bing/Yandex ho rešpektujú)
Crawl-delay: 1

# ── Blokovaní agresívni scrapovací roboti ──────────────────────────────────
# SEO/backlink harvestery: čistá záťaž servera bez akéhokoľvek prínosu.
User-agent: SemrushBot
Disallow: /

User-agent: AhrefsBot
Disallow: /

User-agent: AhrefsSiteAudit
Disallow: /

User-agent: MJ12bot
Disallow: /

User-agent: DotBot
Disallow: /

User-agent: BLEXBot
Disallow: /

User-agent: PetalBot
Disallow: /

User-agent: DataForSeoBot
Disallow: /

User-agent: Barkrowler
Disallow: /

User-agent: SeekportBot
Disallow: /

User-agent: serpstatbot
Disallow: /

User-agent: ZoominfoBot
Disallow: /

User-agent: MegaIndex
Disallow: /

User-agent: SEOkicks
Disallow: /

User-agent: BacklinkCrawler
Disallow: /

User-agent: Riddler
Disallow: /

User-agent: magpie-crawler
Disallow: /

User-agent: trendictionbot
Disallow: /

User-agent: VelenPublicWebCrawler
Disallow: /

User-agent: SurdotlyBot
Disallow: /

User-agent: aiHitBot
Disallow: /

User-agent: Screaming Frog SEO Spider
Disallow: /

User-agent: Sogou
Disallow: /

User-agent: YisouSpider
Disallow: /

User-agent: Mail.RU_Bot
Disallow: /

User-agent: LinkpadBot
Disallow: /

User-agent: SiteAuditBot
Disallow: /

User-agent: netEstate
Disallow: /

User-agent: Mediatoolkitbot
Disallow: /

User-agent: DomainStatsBot
Disallow: /

User-agent: SiteCheckerBotCrawler
Disallow: /

User-agent: AwarioBot
Disallow: /

User-agent: Onalyticabot
Disallow: /

User-agent: grapeshot
Disallow: /

User-agent: ADmantX
Disallow: /

User-agent: Linguee
Disallow: /

User-agent: Sitebulb
Disallow: /

User-agent: seolyt
Disallow: /

User-agent: SputnikBot
Disallow: /

# ── Crawlery zbierajúce dáta na trénovanie jazykových modelov ──────────────
# Obsah portálu je autorský a nie je poskytnutý na trénovanie AI modelov.
# Vyhľadávacie (citujúce) agenty — OAI-SearchBot, ChatGPT-User, Claude-User,
# Claude-SearchBot, PerplexityBot — zámerne blokované NIE SÚ: posielajú
# návštevníkov späť a odkazujú na zdroj.
User-agent: GPTBot
Disallow: /

User-agent: ClaudeBot
Disallow: /

User-agent: Claude-Web
Disallow: /

User-agent: anthropic-ai
Disallow: /

User-agent: CCBot
Disallow: /

User-agent: Google-Extended
Disallow: /

User-agent: Bytespider
Disallow: /

User-agent: Amazonbot
Disallow: /

User-agent: Applebot-Extended
Disallow: /

User-agent: FacebookBot
Disallow: /

User-agent: Meta-ExternalAgent
Disallow: /

User-agent: Meta-ExternalFetcher
Disallow: /

User-agent: Diffbot
Disallow: /

User-agent: omgili
Disallow: /

User-agent: omgilibot
Disallow: /

User-agent: ImagesiftBot
Disallow: /

User-agent: YouBot
Disallow: /

User-agent: Timpibot
Disallow: /

User-agent: Webzio-Extended
Disallow: /

User-agent: cohere-ai
Disallow: /

User-agent: cohere-training-data-crawler
Disallow: /

User-agent: AI2Bot
Disallow: /

User-agent: PanguBot
Disallow: /

User-agent: Kangaroo Bot
Disallow: /

User-agent: ICC-Crawler
Disallow: /

User-agent: img2dataset
Disallow: /

User-agent: iaskspider/2.0
Disallow: /

User-agent: Firecrawl
Disallow: /

User-agent: Bedrockbot
Disallow: /

User-agent: Scrapy
Disallow: /

# ── Sitemap ─────────────────────────────────────────────────────────────────
Sitemap: https://nefro.polascin.net/sitemap.xml